Papers
Topics
Authors
Recent
Search
2000 character limit reached

SPI-GAN: Denoising Diffusion GANs with Straight-Path Interpolations

Published 29 Jun 2022 in cs.LG and cs.AI | (2206.14464v3)

Abstract: Score-based generative models (SGMs) show the state-of-the-art sampling quality and diversity. However, their training/sampling complexity is notoriously high due to the highly complicated forward/reverse processes, so they are not suitable for resource-limited settings. To solving this problem, learning a simpler process is gathering much attention currently. We present an enhanced GAN-based denoising method, called SPI-GAN, using our proposed straight-path interpolation definition. To this end, we propose a GAN architecture i) denoising through the straight-path and ii) characterized by a continuous mapping neural network for imitating the denoising path. This approach drastically reduces the sampling time while achieving as high sampling quality and diversity as SGMs. As a result, SPI-GAN is one of the best-balanced models among the sampling quality, diversity, and time for CIFAR-10, and CelebA-HQ-256.

Authors (2)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (25)
  1. Neural ordinary differential equations. Advances in neural information processing systems, 31, 2018.
  2. Image generation with shortest path diffusion. arXiv preprint arXiv:2306.00501, 2023.
  3. A family of embedded runge-kutta formulae. Journal of Computational and Applied Mathematics, 6(1):19 – 26, 1980.
  4. Taming transformers for high-resolution image synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  12873–12883, 2021.
  5. Learning energy-based models by diffusion recovery likelihood. arXiv preprint arXiv:2012.08125, 2021.
  6. Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30, 2017.
  7. Denoising diffusion probabilistic models. Advances in Neural Information Processing Systems, 33:6840–6851, 2020.
  8. Progressive growing of gans for improved quality, stability, and variation. arXiv preprint arXiv:1710.10196, 2018.
  9. Training generative adversarial networks with limited data. Advances in Neural Information Processing Systems, 33:12104–12114, 2020a.
  10. Analyzing and improving the image quality of stylegan. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  8110–8119, 2020b.
  11. Neural controlled differential equations for irregular time series. In NeurIPS, 2020.
  12. On fast sampling of diffusion probabilistic models. arXiv preprint arXiv:2106.00132, 2021.
  13. The cifar-10 dataset. online: http://www. cs. toronto. edu/kriz/cifar. html, 55(5), 2014.
  14. Improved precision and recall metric for assessing generative models. Advances in Neural Information Processing Systems, 32, 2019.
  15. Improved denoising diffusion probabilistic models. In International Conference on Machine Learning, pp.  8162–8171. PMLR, 2021.
  16. Dual contradistinctive generative autoencoder. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  823–832, 2021.
  17. Adversarial latent autoencoders. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.  14104–14113, 2020.
  18. Improved techniques for training gans. Advances in neural information processing systems, 29, 2016.
  19. Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502, 2021a.
  20. Generative modeling by estimating gradients of the data distribution. Advances in Neural Information Processing Systems, 32, 2019.
  21. Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2021b.
  22. Score-based generative modeling in latent space. Advances in Neural Information Processing Systems, 34, 2021.
  23. Diffusion-gan: Training gans with diffusion. arXiv preprint arXiv:2206.02262, 2022.
  24. Tackling the generative learning trilemma with denoising diffusion gans. arXiv preprint arXiv:2112.07804, 2021.
  25. Differentiable augmentation for data-efficient gan training. Advances in Neural Information Processing Systems, 33:7559–7570, 2020.

Summary

No one has generated a summary of this paper yet.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.